EGH19 v Commonwealth of Australia

High Court of Australia · 2026

EGH19 v Commonwealth of Australia [2026] HCA 7

After the High Court struck down a regulation requiring ankle monitors and curfews for released non-citizens, the government rewrote it around a narrower risk test. Did the rewrite survive Chapter III?

What happened?

In NZYQ the High Court held that the Migration Act provisions authorising executive detention of an unlawful non-citizen whose removal has no real prospect of becoming practicable in the reasonably foreseeable future are invalid to that extent (Gageler CJ and Gleeson J at [1]). The Commonwealth responded with two regimes. In the first, the Minister may grant such a person a Bridging R (Subclass 070) visa, called a BVR, on prescribed conditions, including a monitoring condition (wearing a monitoring device 24 hours each day) and a curfew condition (remaining at a notified address between 10.00pm and 6.00am each day) ([3]). In the second, Div 395 of the Criminal Code (Cth) lets the Supreme Court of a State or Territory, on the Minister's application, make a community safety supervision order against an NZYQ affected person convicted of a serious violent or sexual offence ([4]-[5]).

YBFZ then held that cl 070.612A(1) of Sch 2 to the Migration Regulations, in its original form, was invalid to the extent that it authorised and required the Minister to impose the two conditions ([7]). The regulations were amended to repeal and substitute a new cl 070.612A(1) ([8]). The substituted clause requires the Minister to impose each condition if satisfied on the balance of probabilities that the holder poses a substantial risk of seriously harming any part of the Australian community by committing a serious offence, as defined in cl 070.111, and that imposing the condition is reasonably necessary and reasonably appropriate and adapted for the purpose of protecting any part of the Australian community from serious harm by addressing that risk ([8], [33]-[34]). The rules of natural justice do not apply to the decision to grant the visa subject to either condition, and the holder is invited to make representations only after the decision ([37], [53]). The conditions remain in force for 12 months from the grant, and non-compliance is an offence punishable by imprisonment for between one and five years ([38]).

The plaintiff is a citizen of Papua New Guinea who arrived in Australia as a child in 2000 (Gordon J at [108]). In April 2025 a delegate of the Minister granted him a BVR on conditions that included the monitoring and curfew conditions, and he was released into the community ([110]). He commenced a proceeding in the High Court's original jurisdiction against the Commonwealth, and the parties stated a special case ([9]). The Commonwealth did not seek to reopen the holding in YBFZ that the nature and severity of the detriment imposed by each condition warrants prima facie characterisation of its legislative imposition as penal or punitive ([40]).

What did the Court decide?

The Court answered Yes to Question 1 (the curfew condition) and Question 2 (the monitoring condition), and answered Question 3, who should pay the costs of the special case, with "The defendant" (orders). Five Justices joined in those answers and two would have given the opposite answers. There are six judgments, one of them joint.

Proposition

What is the principle?

Gageler CJ and Gleeson J. Their Honours stated the limitation as one under which a Commonwealth law imposing a detriment of a nature and severity warranting prima facie characterisation as penal or punitive, other than through a court exercising the judicial power of the Commonwealth in adjudging and punishing criminal guilt, contravenes Ch III unless the law is reasonably capable of being seen to be necessary for a legitimate and non-punitive purpose ([14]). They said that the inquiry into necessity has a substantive and a procedural dimension, and "is not simply as to what is authorised to be done but as to by whom and how it is authorised to be done" ([21]). They identified the purpose of the clause as protection of any part of the Australian community from a substantial risk of the harm inherent in particular conduct constituting a serious offence ([46]), and held it legitimate at least in its application to the conduct described in para (b)(i)-(vi) of the definition, without determining the position for all conduct, and described the conduct in para (b)(vii)-(x) as plainly severable ([47]). At the stage of necessity they held that the fixed 12-month duration meant that the detriment "fails the test of reasonable necessity" ([52]) and that "What cannot be accepted to be reasonably capable of being seen as necessary for that purpose is for cl 070.612A(1) of Sch 2 to the Migration Regulations in its current form to authorise imposition of restrictions of that nature other than by a court in the exercise of judicial power" ([56]). They treated the existence of Div 395 as strengthening that conclusion without involving impermissible proportionality analysis ([56]).

Gordon J. Her Honour held that the purpose of the clause "is not a legitimate, non-punitive purpose", because the risk of harm to which it is directed is not sufficiently grave and the cohort to which it applies has no legitimate basis ([63]-[64], [125]-[135]). In the alternative, if the purpose were legitimate, the imposition of the conditions "is not reasonably capable of being seen as necessary for that purpose" ([65], [137]). Her Honour did not reach whether the provision could be read down or severed ([136]).

Edelman J. His Honour held that the conditions are punishment, concluding that abductive inference, or the colloquial duck test, "conclusively points to the home detention and monitoring conditions as being punishment" ([207]). Because they are punitive, no issue arose as to whether the means were reasonably capable of being seen as necessary for a legitimate and non-punitive purpose ([211]).

Jagot J. Her Honour held that the avowed purpose, in the abstract, is capable of characterisation as legitimate and non-punitive, but "the law cannot be objectively and substantively characterised as being for that purpose" ([341]). Her Honour held the law prima facie punitive and, because it was materially over-inclusive, not reasonably capable of being seen to be reasonably appropriate and adapted to its stated purpose, so that it was to be characterised as substantively punitive and within the scope of exclusive judicial power ([341]-[342]). Her Honour grounded that conclusion in five features of the law ([332]-[340]).

The dissents. Steward J, who maintained that the jurisprudence of the YBFZ plurality is not supported by the Constitution, held that he was bound to apply it ([226]). He would not have accepted that the new power is prima facie punitive ([257]), but recorded the defendant's concession that it is ([258]), and held that the power has a legitimate and non-punitive purpose and is not invalid ([270]). Beech-Jones J held that cl 070.612A(1) is valid, applying the reasoning that he had applied in YBFZ ([350]).

Benbrika (No 1). Gageler CJ and Gleeson J described Thomas v Mowbray and Benbrika (No 1) as the two cases in which a Commonwealth law authorising preventive restrictions on liberty outside the constitutional paradigm had been held reasonably capable of being seen as necessary for a legitimate and non-punitive purpose, each by conferring "federal jurisdiction on a court to order preventive restrictions on liberty in the exercise of judicial power" ([24]). Gordon J said that it was important to the Benbrika plurality's reasoning that terrorism "poses a singular threat to civil society" ([126]). Edelman J said: "There are two grave difficulties with the reasoning of the joint judgment in Benbrika [No 1]", the strict dichotomy between punishment and protection and the notion of an "allied Lim principle" ([192]). Steward J, who was one of the Benbrika joint judgment, said that Benbrika (No 1) and Garlett, "whilst not formally overruled, would now appear to have been effectively diluted, or at least re-explained" ([225]). Beech-Jones J said that Benbrika (No 1) "compels the conclusion that cl 070.612A authorises the imposition of the curfew and monitoring conditions for a legitimate and non-punitive purpose" ([374]).

Why does this case matter?

EGH19 is easy to over-read. It is recent, five Justices made the same orders, and it contains several striking remarks about earlier cases. In an essay or problem question the usual error is to take one of those remarks and present it as what the Court decided, so that the citation attributes to seven Justices something that one or two of them wrote.

The case is also a worked example of a legislative response to an adverse High Court decision. A redraft that answers the points made against the earlier law is not thereby safe, and a student asked about a redrafted law has to compare the new text with the reasons given against the old one, not only with the headline result.

Exam and application relevance

For the orders, cite the Court: Questions 1 and 2 were answered Yes by Gageler CJ and Gleeson J, Gordon J, Edelman J and Jagot J, with Steward J and Beech-Jones J dissenting.

For a reason, attach it to the Justice or Justices who gave it, with the paragraph. Where an answer relies on a passage from a dissent, say that it is a dissent.

The questions stated in the special case concerned cl 070.612A(1) of Sch 2 to the Migration Regulations, so Div 395 of the Criminal Code and Div 105A were not the subject of the questions the Court answered. Remarks in the judgments about court-ordered regimes are the remarks of the Justices who made them.

Check your understanding

1. What were the Court's answers to Questions 1 and 2, and which Justices would have answered No? 2. Whose reasons hold that the purpose of the substituted clause was not a legitimate and non-punitive purpose, and what do Gageler CJ and Gleeson J say about the same purpose? 3. Whose reasons hold that the conditions are punishment and for that reason leave the necessity of the means unexamined? 4. Steward J was one of the Benbrika joint judgment. What did his Honour say about Benbrika (No 1) and Garlett in this case, and did his Honour join the orders? 5. What did Gageler CJ and Gleeson J say that Thomas v Mowbray and Benbrika (No 1) had in common?